Previous to now it had always relied on a custom patched gcc
and a custom sim. In the interim since the initial port went in
some time in 2015 GCC and QEMU have both officially picked up support
for the architecture and the machine that was emulated in the previous
emultor.
Using gcc 10.2 fix up the build and get it basically working. Timers
seem to not be working right but it's probably fairly easy to fix.
Just an itch that has bothered me for a while. Current time is not
strictly speaking platform specific though it's usually associated
with platform code. Move it into its own header and add some text
to explain a bit whats going on.
Add some more comments for other headers while in there.
Using the RTC1 module for the LK tick (instead of arm systick)
Use the nrf_hal library for peripheral manipulation instead of
direct register access. Fixed race condition in current_time_hires
which would result in non monotonic readings.
Tested with clock_tests on:
nrf52-pca10040-test (nrf52832)
nrf52-pca10056-test (nrf52840)
1) Decode FSC and dump more human readable status
2) Add support of stack unwinding as referred from
arm64 procedure call standard and frame pointer usage.
3) Compiler options for not omitting frame pointer
are enabled to ensure usage of frame pointers even
with higher optimization levels enabled.
Signed-off-by: vannapurve <vannapurve@google.com>
Since the i2c driver is optional (not required for platform) the
i2c_init should be done during target initialization. The flags
used to include the i2c driver should be in either target or project
files so it makes sense that the i2c_init should also be called from
target files.
When the nrf52xxx platform was changed to use the Nordic nrfx
libraries the pca10040 target was broken due to differences in
how the platform handled initialization.
In a previous patch, on architectures where SMP is disabled,
thread_pinned_cpu() will always return a negative number. GCC 10.2
is somewhat more aggressive about negative shift values so it wasn't
detecting that the second path of an if statement is never taken.
Solution is to cache the pinned cpu value and use it for both the
test and the operation.
This change ensures that during thread scheduling
only needed cores are woken up as per the thread's
CPU affinity rather than all cores. This would avoid
un-necessary wakup for other cores from lower power
states without having to execute anything.
Signed-off-by: vannapurve <vannapurve@google.com>
All shipping ARMv7-R processors include VFPv3-D16, a subset of VFPv3
with only 16 double-precision registers. The first -R profile CPU
with a complete VFP or NEON implementation is the Cortex-R52,
implementing ARMv8-R.
LK's context switch code had a dynamic check and would only save
d16-d31 if present, but gcc/gas will not assemble code that includes
references to d16-d31 when mcpu=cortex-r4f or other v7-R CPUs.
Ideally we'd key the #if off of __TARGET_FPU_VFPV3_D16, but that
only appears to be defined by the ARM compilers, not gcc. Use
V7-R as the key instead.
Using Nordic's nrfx driver for the clock control peripheral since
it address a handful of errata and abstracts some differences
in the nrf52 family of parts.
Platform gpio driver now uses the nrfx library for gpio manipulation.
This allows utilization of second gpio bank present on some parts
and handling of errata.
Use twim(i2c) driver from nrfx library. See comments and patterns in
target/nrf-pca10056 for info on how to properly utilize driver as it
requires some GLOBAL_DEFINES and gpio defines.
From working with -Wpacked its clear that some of the virtio
structures are overly packed and result in bad codegen on particular
architectures such as riscv. Roll back from using packed but
statically assert that the sizes are correct for some future arch
where things naturally pack differently.
Mostly just a few warnings where things are promoted via passing floats to
printf. Those we should generally remove anyway because they're just
benchmarking code. Most things LK runs on either doesn't have float or
doesn't have double sized floats.
Some of the structures, notably 'cmd', in the lib console stuff are a
little too generically named and have collided with some other code
so prefix the names a bit more cleanly with console_
The change is largely mechanical, and folks with out of tree code can
easily switch by renaming:
cmd -> console_cmd
cmd_args -> console_cmd_args
cmd_block -> console_cmd_block
console_cmd -> console_cmd_func
Apologies if this breaks you but it should be pretty easy to fix.
As specified in ST reference manual RM0091, the prescaling factor
applied by timers is 1 plus the value in the corresponding prescaler
register. (See section 17.4.11 for TIM1, and analogous sections for
other timers.)
This change causes stm32_timer_capture_setup to take the prescaling
factor as input rather than the register value.
-Stop using 512 bytes of bss and use a proper printf output
routine if the user wants to override it.
-Add a test case for unittest itself to see what the failure printfs
look like and make sure the test registration is working.
Refactor the two linker scripts to put .sdata at the end of the data segment
and .sbss at the start of the bss segment to try to maximize the reach of
the global pointer.
Also generally tries to clean up and align the two linker scripts as a new
golden standard for other architectures.
Can't remember precisely what this fixes to be honest, but I
remember there was some sort of recursion on one of the targets on
one of the versions of the compiler.
No need to malloc this memory if it's always going to be initialized
anyway. Add history disabling switch to a few targets to make sure both
builds are tested.
Nothing was using console_register_commands so just move the command
block into a read only section. Remove the linked list stuff but add a
block name to print in its place.
Mostly deduplicating x86-32 and x86-64 code since virtually all of it
can be shared.
Fixed up some cpuid usage which was not properly marking registers as
clobber.
Been carrying this flag around for years but from sleuthing around in
the compiler it seems to only exist as the opposite to -fno-inline which
has an actual effect. Only reason -finline would do anything would be to
cancel a previous -fno-inline switch.
From looking at the dissassembly the compiler wasn't doing a good job
with the overly complicated version shared with the larger arm cores.
Simplify it by recognizing that you can save the state directly from
PRIMASK and restore it more simply.
